College Choice | Hacker News Reader

College Choice

github.com

122 pointsdanso26 comments

Some context:

This is the github repo for this site: https://collegescorecard.ed.gov

If you want just the data, here's the landing page with documentation: https://collegescorecard.ed.gov/data/

Direct link to the zip: https://s3.amazonaws.com/ed-college-choice-public/CollegeSco...

The NYT recently wrote an analysis of the salaries data, finding that "at some expensive colleges, the salaries of students 10 years after enrollment are bleak, and there is an earnings gender gap at every top university":

http://www.nytimes.com/2015/09/14/upshot/gaps-in-alumni-earn...

18F is the GSA team that is building a lot of the federal websites and APIs and open sourcing them.

According to this data, University of Phoenix grads make more on average than UT, Texas A&M, Texas Tech, etc. grads. Something smells fishy.

Here's the data file they used. The University of Phoenix only has 72 data points. https://s3.amazonaws.com/ed-college-choice-public/Most+Recen...

I can't find anything about methodology other than the fact that only students who receive federal financial aid may be included.

I'll guess the data is skewed because University of Phoenix caters to people who are already working. Work experience + degree is better than just a degree.

That was my thought too. It would be interesting if there was some measure of salary increase upon graduation.

I suspect that all state universities in locations that have cheaper cost of living that the national average (Texas, for instance) are hampered by viewing the data in the way it is being shown.

As a Texas Tech grad, I'm not that shocked.

I'm an undergraduate who made this decision in the past 1000 days. I went to the website and tried to see how much it would have helped me. It wouldn't have. Here are my criticisms.

* This tool should be clear that it's geared towards 17 year olds that are mature enough to be able to view college as a financial investment. That's not to say that college can't be chosen for the experience of it as well, but I don't think you can recommend colleges on such a personal variable. This type of site can be a great tool if you're willing to put that aside for a moment and merely get information about potential return on an investment, which is useful for everyone regardless of how they feel about what college ought to be.

* Factor in how confident a student is in what they want to study. I think the largest cost sink is in people who don't know what they want to study, but decide to go to an expensive school and say that they'll figure it out in their first two years. What happens when you don't, or school ends up being hard, and you have friends made and leases signed and now you've locked yourself into paying $80k for something you're not going to see return on? Would you have made that decision before?

* Average salary after school is an awful metric. You can't group engineers with those that study humanities and then present that as a meaningful statistic. It makes the person studying philosophy feel more confident in spending money, and it makes the person studying engineering more likely to deviate from liberal arts schools, when in actuality a school can accommodate both groups well but with somewhat different outcomes.

* How do you run a site like this and not ask what their SAT / ACT scores are? This is most certainly the most predictive factor for where you'll go.

* How do you not ask where they live, and talk about in-state tuition rates?

Ultimately it's a nice idea with poor execution, but it's close.

> How do you run a site like this and not ask what their SAT / ACT scores are? This is most certainly the most predictive factor for where you'll go.

Unfortunately the actual data (at least that the govt has) on this is terribly incomplete. With a quick look (this may be off a bit) 6422/7598 colleges from their set are missing this data.

I agree with most everything you said, though.

Hi Nick,

Agreed! There are lots of places where this tool could be improved! (Disclaimer: I work with people at 18F, Dept. Of Ed, USDS who built this tool, but I wasn't involved and don't have any privileged information and so here are some of my personal thoughts)

If I can add a bit of color:

1. This was developed open-source on GitHub. That means you're welcome to file an issue or comment at https://github.com/18F/college-choice/issues (Given the interest and number of issues filed so far, I'll probably take this opportunity to say that it might be a while before the team gets back to you and appreciate your help and patience!). This shouldn't impress you, but it is, because it's a complete paradigm shift from how government typically works.

2. Average salary is not a great metric and unfortunately, as far as I can tell, the data isn't sliced by major. You're right in your analysis of how it skews the data. I actually think loan repayment is probably the more interesting metric - have others in your income class been able to repay their debts? You're also right that this tool doesn't help people see that choosing college A over B because of a $X,000 delta in average salary isn't a good idea as it's subject to all sorts of personal decisions.

3. This is really cool not because the tool is perfect, but because the government is releasing an API to go along with this - a number of private companies and non-profits have already integrated the data into their tools. This data, which hasn't been possible to get before, can now be integrated into the tools that were helpful to you (the ones that calculate in-state tuition, and ask you for sat/act scores). The underlying API also has many more data fields than the tool exposes. If fact, you can build one yourself! (https://api.data.gov/signup and https://collegescorecard.ed.gov/data/documentation/)

> as far as I can tell, the data isn't sliced by major

In reading some of the surrounding commentary (as an 18F'er who was only tangentially involved with this project as it also posting personally), it looks like data by major isn't available. Same thing applies to data for all students (vs only those who get aid).

And ditto that you should file an issue on GitHub!

PS: Hi!

I typed this comment up as a quick break from work, and rereading my comment I can say I did not have enough editing. I didn't mean to come across so critical. It's just so close to being a great idea that can solve what I see as a huge problem-- why are people taking out loans for a bullshit University of Phoenix degree? Because they don't know what their options are, and they don't know what it's going to cost, and they have some feeling in the back of their head that college is always a worthwhile investment. That's how we nationally end up with more student debt loan than credit card debt.

With a little more data and a bit more prodding, this could genuinely fix that. No longer will people be getting their primary source of information from 2:00AM University of Phoenix commercials.

When I said the execution was poor it was not to say that this is somehow botched, it's just that it's so damn close. I think 18F is really cool, and I hope it continues to have cool projects.

At the same time, the tool genuinely wouldn't have helped me, and on HN I feel like that type of criticism is useful. Not just for the poster, but for people who are interested in solving problems in this realm and want to get in the head of someone who would use it.

PS: I signed up for an API key. Thanks for the comment. I'll poke around this weekend.

I disagree with your first point. You can't view education as simply financial investment vs 'experience.' It's also a personal investment. By that I mean one of the most important things that happens to people in this group is that they grow up and you want to have an environment that allows this to happen without getting in the way of your education.

Will going to this school allow you to be a leader? Will you be supported in your choices without being punished for failure? This is as much determined by your peers as the faculty, and it's pretty important. And I don't think 17-year-olds are mature enough to understand this; if they did, they wouldn't need to go to school to learn it.

There needs to be a better-than-black-and-white way to use salary as a measuring stick. Perhaps salary above the local mean, or normalized by cost-of-living indicators, or something. As it is, if you earn $100k/yr in pretty much 90% of the country, you're doing fantastically and probably one of the wealthier people in town, but if you live in the top major metros (LA, NYC, Boston, SF/SV, DC, Seattle, Chicago, ...) you're probably wishing you had a partner who earned a similar salary so you could even consider buying a house some day. Just living within 5 miles of a Whole Foods doesn't mean someone is necessarily happier or better off.

I agree, but I don't think you can get too specific with any metric without imposing your own value system on everyone. For example, some believe SF/SV is overpriced, some view the cost of living as a luxury expense and think it's easily justified by the weather/culture/etc.

Agreed - it's an unfortunate that such an objective measure doesn't exist. I like to think of the problem as SF (for example) not having its own currency when it does have its own economy. SF dollars are worth less than Kansas dollars, for example. If we had a way of breaking this down, it might be easier to see which parts of pricing differential are "luxury expense."

People flock to highly selective colleges from all over, and go all over when they're done. You can't normalize cost of living to the location of the school, you'd have to look at each graduate's salary compared to average in their field where they ended up. I would love to see that data, but it would be a lot of effort to gather.

I've been asked before by younger relatives for advice about college, and it's really hard to give. I only went one school, I don't even know how it would have been different for me if I had gone elsewhere, much less how to translate that into advice for someone else.

I haven't thought about it too much, but it occurs to me that maybe finding people who were unhappy with their experience would be more useful than finding people who were happy.

It's too bad all the 'average cost' data carries the caveat that it only accounts for the cost for in-state students. That makes the cost part of it somewhat useless unless you are limiting your search to in-state schools only.

Agreed. And this makes a world of difference for some institution. Georgia Tech for example charges more than 3 times more for Out of state students[1] while scorecard shows an average of $11k [2]

[1]: http://factbook.gatech.edu/student-related-information/tuiti...

[2]: https://collegescorecard.ed.gov/school/?139755-Georgia-Insti...

I feel that the fact that they take average aid into account can also be a bit misleading. Harvard is just $3K more than Georgia Tech? [3] Probably be best to show Full tuition, then average aid by income levels.

[3]: https://collegescorecard.ed.gov/school/?166027-Harvard-Unive...

I think that the data available here is a good start. To make an informed decision about which school to attend, it would require the students to identify the factors that are important to them. Check out the link below for an example of how one might choose the right school for themselves. https://paramountdecisions.com/education.html

What does "% Earning Above HS Grad" mean? This sorting option doesn't seem to correlate with any of the bars "Average Anual Cost", "Graduation Rate" or "Sallary After Attending".

It means the percentage of attendees who earn more than the average income of a high school graduate.

Ah, so the idea is that universities like MIT and Harvard with very high post graduation salaries, really only have so in average, and many people earn much less. At least to me it is quite surprising that those two statistics are so different.

It's this hosted somewhere?

Top of the page https://collegescorecard.ed.gov/

Reply on news.ycombinator.com